Anomaly Detection giving (Blank) expected value

[rog111]

Hi,

I've been playing around with the Anomaly detection recently.  I'm trying to understand when anomoly detection can't detect anomalies.

Here's the example I'm playing with.  Visually, it looks like there's an anomaly.  

[TheoC]

Hi @rog111 

In your data, would you view the anomaly as the spike  in population, or is the anomaly is the blanks presented in the tooltip?  

Cheers

Theo

[rog111]

I would view the spike as the anomaly as .  At the same time, I'm wondering why expected values could not be calculated.